Effect of mixing section configurations on combustion efficiency of Mg-CO2Martian ramjet

Xu WANG, Ynpeng BU, Xu XU, Qingchun YANG,*

aSchool of Astronautics, Beihang University, Beijing 100191, China

bShenyuan Honors College, Beihang University, Beijing 100191, China

KEYWORDSCombustion efficiency;Magnesium-base fuel;Martian ramjet;Mars exploration;Structural configurations

AbstractExperiments were conducted to determine the effects of the mixing section configurations on the Mg-CO2Martian ramjet combustion efficiency.It was carried out at a mainstream mass flow rate of 110 g/s and a temperature of 810 K.The chamber pressure was measured under different configurations and Oxidizer to Fuel (O/F) ratios.Results showed that the engine achieved selfsustaining combustion and worked stably during experiments.The pre-combustion chamber is needed to increase the combustion efficiency and promote the full combustion of the powder.After the configuration of the pre-combustion chamber, the best combustion efficiency reached 80%when radial powder injection and lateral carbon dioxide intake were used.In addition, the O/F ratio in the pre-combustion chamber decreased from 0.67 to 0.31, resulting in an 8% increase in the combustion efficiency.It was speculated that different mixing section configurations and the variations in an O/F ratio within the pre-combustion chamber impacted the combustion efficiency and in essence, all affected the flow velocity and residence time of the two-phase flow in the combustion chamber.

1.Introduction

Mars exploration has witnessed a new boom over the past decades1–3.However, the significantly increased landing payloads are beyond the capability of currently available vehicles,which are primarily restrained by the large distance and prolonged duration of interplanetary space travel.Economically,it is necessary to develop a new technology to solve this problem.Recently, In-Situ Resource Utilization (ISRU) for Mars has been proposed, and it is desirable to have fuel for a Martian vehicle available on Mars4.The atmospheric breathing ramjet is an attractive propulsion system since it enables the Martian atmosphere as the oxidizer for thrust augmentation5.More importantly,in contrast to a turbine engine,the structural simplicity of a ramjet engine suggests higher reliability.The Martian atmosphere is 95.3% carbon dioxide.Any working medium capable of reacting with carbon dioxide and igniting can be considered a possible propellant.

Recent studies have found that metal fuels offer high specific impulses and are easily stored.The main metal fuels that can be burned in the carbon dioxide environment are aluminum, lithium, magnesium, and beryllium.Despite having the highest specific impulse, beryllium oxides are exceedingly poisonous and carcinogenic6,7.Aluminum and lithium also have a high specific impulse, but the reaction intensities with carbon dioxide are both unsuitable.Fortunately, magnesium has a high specific impulse and a good reactivity in carbon dioxide while remaining non-toxic.Magnesium oxide, in particular, was discovered to make up between 8.3% and 8.6%of the Martian regolith.As a result, the combination of magnesium and carbon dioxide has the greatest potential as a ramjet propellant alternative.At present, Mg-CO2combustion mainly focuses on the single-particle combustion mechanism8.Previous research has focused on the effects of pressure, temperature9, products10–11, particle size12–16, and pulsating combustion17on Mg-CO2combustion.Meanwhile, the ignition delay time and burning time of magnesium particles in carbon dioxide are also of major concern5,15,18.While various experiments on Mg-CO2rocket engines and jet engines have been conducted, far less has been reported on Martian ramjet10,19–21.These studies of rocket and turbine engines also provide a relevant reference in the development of Mg-CO2ramjet engines, and the theoretical viability of the Mg-CO2Martian atmosphere-breathing propulsion system has been demonstrated5,22,23.Recent years have witnessed the development of the theoretical preliminary design for the magnesiumfueled Martian ramjet engine and initial performance estimations of a magnesium-fueled ramjet cycle.However, research in such a ramjet is still in its infancy.

This study aims to explore the feasibility of Mg-CO2ramjets using experimental methods.Simultaneously, the effects of the ramjet’s mixing section configuration and the O/F ratio in the pre-chamber on the combustion performance merit further study.The detailed investigation includes the combustion characteristics of the ramjet and the effect of various structural design parameters on the Mg-CO2combustion process,including powder injection pattern and inlet of high enthalpy carbon dioxide.On this basis, the combustion efficiency was calculated and analyzed in detail.

2.Experimental details

2.1.Experimental system

Fig.1 depicts a schematic diagram of the experimental system used for the Mg-CO2ramjet.The direct-connect experimental system consists of a powder dispersion, electric heating, and a ramjet combustor.

Fig.1 Schematic diagram of Mg/CO2ramjet experimental system.

The powder dispersion system may inject a constant mass flow rate of magnesium powder into the combustor uniformly and steadily to participate in combustion via the combined action of carbon dioxide gas carrying and piston pushing.Meanwhile, a 75 kW electrical heater raises the temperature of the mainstream to approximately 810 K,simulating the high enthalpy mainstream of CO2at the exit of the inlet during hypersonic flight.Compared to other types of heaters,the electrical heater has the advantage of a moderate heating rate and the absence of pollutants.Additionally, to simulate the gas throttling and control the mass flow rate of the mainstream,a critical nozzle was used to substitute the inlet throat.Hence,the mass flow rate of mainstreams is exclusively determined by the total pressure of the gas above the throat.When illustrated in Fig.2, the speed of high-enthalpy mainstream drops to around 50 m/s as it enters the mixing section, where the powder is gradually dispersed.However, the configuration of the mixing section has a significant effect on the full powder mixing, which directly affects combustion efficiency24,25.In addition, the O/F ratio in the pre-combustion chamber and powder injector plays a crucial role in ignition and efficient combustion in the Mg-CO2propulsion system11.Thus, three mixing section configurations of the Mg-CO2ramjets and different O/F ratios in pre-combustion chamber were investigated in this study.The difference between the three configurations is whether to configure the pre-combustion chamber and the position of the mainstream carbon dioxide inlet.Configurations A and B mainly tested the difference between axial and radial intake.Configuration C explored combustion characteristics without a pre-combustion.

Fig.2 Different mixing section configurations (except for varying configurations of mixing section, other structures in three constructions are identical).

2.2.Mixing section configurations

Obviously, the primary distinction between the three configurations lies in the layouts of the precombustion chamber and main inlet, as illustrated in Fig.2.Other main structural parameters of the three configurations are listed in Table 1 and are basically consistent.The structural parameters of the Mg-CO2ramjet combustion chamber design references can be found in Refs.5,23,26.

In terms of spatial distribution, the flameholder divides the combustion chamber into two zones.A mixing section is located upstream, while a secondary combustion section is located downstream.Magnesium powder and high enthalpy mainstream gas were mixed in the mixing section and the initial combustion occurred upon torch activation.Fig.3 provides an overview of the inner flow field in the combustor,illustrating how the ‘‘V-shape”flameholder (which is widely used due to its structural simplicity and convenience)can form a localized low-speed reflux zone in the combustion chamber to maintain the stable magnesium powder combustion in the mainstream of CO2.

2.3.Magnesium powder

Magnesium powder in the present experiments is composed of spheroidal or nodular magnesium particles with a mean volume diameter d30of about 64.6 μm.Magnesium powder of the same size was previously employed in the combustion experiments with single particles in the atmosphere of carbon dioxide or its mixtures with argon.Experiments with particle geometry demonstrated that magnesium particles with a radius of 63 μm provided good ignition and a smooth flow through the powder dispersion system6.The Scanning Electron Microscope (SEM) image of the powder, as shown in Fig.4, illustrates information on the particle size and morphology27.Additionally, Fig.4 also shows the particle size distribution obtained using a laser particle size distribution instrument.

2.4.Powder dispersion system and gas flow

In order to feed the magnesium powder stably and precisely,a powder dispersion system based on the previous investigations was designed28,29.The powder dispersion system in Fig.5begins with loosely packed powdered magnesium powder in a cylinder(30 cm long,and 4.0 cm in diameter).At the bottom of the cylinder, an electric piston propels the powder upward at a velocity of between 0 mm/s and 20 mm/s.When the powder reaches the top of the cylindrical powder storage tank,the high-velocity carbon dioxide gas (300 K) is sprayed into the tank through a combined throttle device comprised of a 2500 mesh strainer and a 0.2 mm circular slot.The carrier gas reaches critical sound velocity due to the combined action of the strainer and circular slot.The transverse high-velocity jet impinges on the powder’s surface and disperses the powder into the carbon dioxide carrier gas by shear action.

Table 1 Main structural parameters in combustor.

Fig.3 Flow field in combustor of Configuration C.

Fig.4 SEM photograph (up) and volumetric particle size distribution (down)27.

Fig.5 Schematic diagram of powder dispersion system.

Moreover, when the powder fuel is choked, the pressure difference between the powder storage tank and the combustion chamber is large enough to change the combustion chamber pressure, which has little effect on the injection pressure.This means that the powder dispersion system can deliver a steady mass flow rate of powder, while the mass flow rate of powder injected into the combustion chamber is solely determined by the piston speed.As illustrated in Fig.6 and Fig.7,the results of the powder injection test show an excellent linear relationship between the mass flow rate of powder and the piston velocity.The mass flow rate of the powder is determined while it is being sprayed into a container by real-time measurement of the total mass with an electronic balance and the displacement of the piston using a displacement sensor.Calibration findings indicated a maximum error of 4.7%.Notably, the powder concentration is mainly determined by the piston speed and mass flow rate of carrier gas.During combustion experiments, powdered metal in gas suspension flows up from a conical diffuser into a 1 m tube with an inner diameter of 6 mm.Finally, the two-phase flow is injected into the mixing section through a 2 mm powder-CO2injector.

2.5.Experiment conditions

The experimental conditions are designed and shown in Tables 2 and 3 in accordance with the mixing section configurations in Section 2.1.The cruise parameter of the aircraft is designed to be 4.0 Ma.The flight ambient pressure is 484 Pa,and the ambient temperature is 235 K on Mars.Therefore,the total temperature of the incoming flow can be calculated.Limited by experimental conditions, the maximum mainstream mass flow rate was 110 g/s and the powder mass flow rate remained constant at ˙mp=20 g/s in all the test cases of this study.

Fig.6 Relation between total mass of extruded magnesium powder and piston displacement.

Fig.7 Mass flow rate of magnesium powder vs piston velocity.

Table 2 Parameter of mainstream and carrier gas.

To ensure a sustained reaction in the combustion chamber after deactivating the torch, a higher temperature in the precombustion chamber and upstream of the mixing section is preferable.The effects of combustion temperature and powder spraying speed were considered, and the oxygen/fuel ratio should be close to exactly the oxygen/fuel ratio, as shown in Fig.8.In addition, it is better to burn in a fuel-rich state.Hence,O/F ratios in the pre-combustion chamber and the exit of the powder-CO2injector were designed to be 0.3 and 0.6,respectively.When stable combustion can be maintained upstream in the combustion chamber, the O/F ratio should be increased as much as possible in the secondary combustion section to optimize the specific impulse performance and reduce the deposition of condensation products in the combustor.It should be noted that when the combustion temperature is lower than 2000 K,carbon deposition occurs in the combustion chamber.Carbon deposition in combustion will be detrimental to the engine’s reusability.Thus, the influence of temperature and mass flow rate on combustion is considered.The O/F ratio in the secondary combustion section was designed to be close to 6.0.

2.6.Data processing methods and uncertainty analysis

In this paper, combustion efficiency is used to evaluate the ramjet performance.The combustion efficiency of engines is defined below28:

where Pcis the average chamber pressure of the combustion chamber during steady engine operation, Atis the area of the nozzle throat,and ˙mtotalcontains the mass flow rate of carrying gas,powder and mainstream carbon dioxide.This paper used sonic nozzles to regulate the mass flow rate of all gases.The following Eq.(4) illustrates measurement principles of the mass flow rate of gases:

Table 3 Experimental conditions for different mixing section configurations.

Fig.8 Mass fraction of condensate products and temperature of combustion with different O/F ratios.

where At;cis the nozzle throat area; K is a constant related to the specific heat ratio of gases.The mass flow rate of gases ˙m can be calculated by measuring the total pressure Ptand temperature Ttof sonic nozzles.

All pressure data were measured using the CYB-20S thin film sputtering pressure transducer of Beijing WESTZH company.The accuracy of all pressure measurements is ±0.25%FS (Full Scale).Sheathed thermocouple measurements were used in the temperature data with a measurement accuracy of±1 K.In addition,the accuracy of mainstream temperature control is approximately±10 K by the electric heating system.In data processing,the uncertainty propagation formula of the error propagation theory30is shown in Eq.(5):

where E（xi）denotes the relative error of xi.The relative error in combustion efficiency is also significant for the performance analysis.Combustion efficiency is determined by the actual and theoretical characteristic velocities.The relative error of experimental characteristic velocity is determined by Eq.(6):

It is proportional to the relative error of the ramjet chamber pressure Pc, the nozzle throat diameter Dt, the oxidizer, and fuel masses.The relative error for the chamber pressure is dependent on the sensor sensitivity, maximum sensor operating pressure of 0.5 MPa and pressure dynamic fluctuation(ΔP), which can be found in Eq.(7).

The relative error of gas mass flow rate is mainly caused by pressure,temperature,and calibration errors(the errors of the flow controller throat diameter Dt;cand coefficient K are equal to those of the mass flowmeter).On the other hand,the relative error of theoretical characteristic velocity is mostly affected by the error of the O/F ratio(EOF)and the non-dimensional slope of the theoretical characteristic velocity in relation to O/F(f′（OF）/f（OF）).

3.Results and discussion

3.1.Experimental results

The typical chamber pressure and injection pressure versus time under Case 3 were plotted in Fig.9.After the mainstream is heated to 810 K, the suspended magnesium powder is sprayed into the pre-chamber through the powder-CO2injector.Magnesium powder and mainstream are mixed in the mixing section.After establishing a constant injection pressure, a gas-oxygen/kerosene torch is activated and operated for 3 s to rapidly raise the combustion chamber’s temperature.

Due to the ignition temperature of the torch(2100 K)being significantly higher than the ignition point of magnesium, a certain proportion of magnesium powder reacts with carbon dioxide and bursts into flame.Meanwhile, as shown in Fig.3, the fuel gas will form a low-speed backflow zone of the flameholder,further stabilizing the combustion.Therefore,after the torch is activated, the magnesium powder remains stable in the mainstream of carbon dioxide, and the engine chamber pressure remains stable at 0.29 MPa.The engine worked stably for 4 s under self-sustaining combustion conditions,and then shut down when the powder-CO2injector valve was closed.Likewise, combustors operating under any of the other working conditions in the previous section can work normally and maintain a stable combustion chamber pressure.The combustion chamber pressure curve and the normalized chamber pressure are displayed in Fig.10.These experimental tests preliminarily proved the feasibility of the Martian Mg-CO2ramjet.

Fig.9 Typical chamber pressure and injection pressure curve in Case 3(containing two pictures of flames with or without a torch).

Fig.10(a) shows that the chamber pressure in the four experimental conditions except Case 2 reaches a steady-state soon after the torch.Although Case 2 can achieve selfsustaining combustion, its combustion chamber pressure does not reach a steady state until 1 s before the powder supply stops.This phenomenon is caused by a sudden pressure drop in the test environment.

In order to discharge the exhaust gas in the experimental chamber in time, the air ejector behind the engine nozzle was started in Case 2.Due to the sudden drop in the combustion chamber pressure just after the torch work,the gas at the nozzle throat has not yet reached the critical speed of sound, as shown in Fig.11.Therefore, the engine is in a pressure build-up process until a stable chamber pressure is reached.Here in this study, the last period of stable combustion was selected as the parameter selection period of the normal working stage of the engine.From the perspective of engine combustion stability, Case 1 has the worst stability.This may be related to its lower combustion temperature and severe condensate deposition in the engine.In addition, for engines like Case 4 without precombustion chambers, the chamber pressure fluctuates greatly during startup.Therefore, from the ramjet working stability and ignition characteristics, it is necessary to use axial powder injection and pre-combustion chamber configuration.

3.2.Analysis of performance and deposition

The combustion performance of the Mg-CO2ramjet was thoroughly analyzed under different experimental conditions, and the engine’s combustion efficiency is depicted in Fig.12.

Fig.10 Combustion chamber pressure curves and normalization.

Fig.11 Combustion chamber pressure and test ambient pressure in Case 2.

The main difference between Case 1 and Case 2 in mixing section configurations is that the position of the powder-CO2injector and the mainstream inlet was swapped.Case 2 is 6%more combustion efficient than Case 1 due to the larger backflow zone created by the injection method, allowing the powder to be more fully mixed with carbon dioxide.Moreover,radial injection of powder results in an apparent deposition in the impact area of the mixing section due to the combined action of the torch.The accumulation of unburned products formed in the mixing section,as shown in Fig.13,is primarily composed of unburned magnesium particles, magnesium oxide, and periclase on the surface.A large number of condensed products in the combustion chamber will seriously shorten the engine’s working life.Experimental results suggest that axial powder injection and the mainstream inlet lateral layouts such as Case 2 should be adopted.

Fig.12 Combustion efficiency under different configurations and O/F ratios.

While large deposits are avoided in Configuration B,increased engine combustion efficiency requires further reduction of the average powder injection velocity to allow the magnesium powder to burn completely in the combustion chamber.The main reason is that the Damkohler number can be effectively increased by reducing the injection speed when the engine structure and powder size remain constant.Eq.(11) can be used to define the Damkohler number:

where τsis the residence time of Mg particle and τchis the chemical reaction time.L,uj,τiand τbrepresent the characteristic length, average powder injection velocity, ignition delay time and burning time, respectively.

Therefore, in Cases 3 and 4, the mass flow rate of the carrying gas was reduced to slow down the average powder injection velocity, and the O/F ratio was dropped from 0.6 to 0.3.Its equivalent ratio increased from 3 to 6 in the combustor.The experimental results indicate that the combustion efficiency of Configuration B rose from 72% (Case 2) to 80%(Case 3).Nevertheless, this change also reduces the O/F ratio in the pre-combustion chamber.Due to the structural similarity of the mixing section to the extended section of the precombustion chamber,the deposition in the two sections is similar.Only the deposition of the mixing section and the flameholder position are shown in Fig.14.

It was found that a lower O/F ratio results in carbon deposition in the combustor, and a significantly increased deposition in the flameholder due to the decreased combustion temperature by comparing the deposition under different O/F ratios.In contrast,a higher O/F ratio leads to condensation products that are mostly manifested as the white magnesium oxide sheets in the pre-combustion chamber and mixing section.The logical question is whether the deposition can be further reduced by removing the pre-combustion chamber while maintaining the same O/F ratio.

Fig.13 Deposition in impact area of mixing section.

Fig.14 Deposition of mixing section and flameholder in Configuration B.

Unfortunately,the experimental results disproved this conjecture.Without the pre-combustion chamber,the combustion efficiency of the combustor decreased by 20%, and chamber pressure was reduced by 25%.Additionally, numerous condensation products are deposited on the surface of the nozzle’s convergence section.According to the X-Ray Diffraction(XRD)test results,the main components were unburned magnesium powder, magnesium oxide, and periclase, as shown in Fig.15.The cause of this phenomenon could be that the powder stagnated in the combustor for an insufficient amount of time, and was not evenly mixed with the mainstream in Configuration C without a pre-combustion chamber.At the same time, as the initial temperature of metal powder injected into the mixing section was lower,it took a long time for the metal powder to reach the ignition point.As a result of the previous findings,it can be concluded that a pre-combustion chamber is necessary for efficient and stable combustion in a combustor using magnesium particles of this size level.

Fig.15 XRD analysis of deposition in combustor without pre-combustion chamber.

4.Conclusions

The magnesium and carbon dioxide combustion experiments were successfully conducted on a direct-connected ramjet combustor.To obtain ramjet performance data under different conditions, experiments were conducted at a mainstream temperature of 810 K and a mass flow rate of 110 g/s, and the engine achieved self-sustaining combustion and worked stably during all test cases.The following conclusions can be drawn from the results obtained:

A direct-connected ramjet experimental system for Mg-CO2was successfully established,and three mixing section configurations were designed.Powder supply and mainstream heating tests proved that these supporting systems achieved the expected function.Engine test results show that the Mg-CO2ramjet is feasible for a Martian atmosphere-breathing propulsion system.Under the experimental conditions of this paper,the highest combustion efficiency achieved is 80%.

Mixing section configuration has a significant influence on Mg-CO2ramjet performance.The pre-combustion chamber is necessary for ramjet engines using magnesium particles at 64 μm levels.Under the same conditions, the combustion efficiency increases from 60%to 80%with the addition of the precombustion chamber.Moreover, radial powder injection and lateral carbon dioxide intake are beneficial to fully mixing the powder with the mainstream while avoiding severe deposition in the combustor.It was speculated that properly reducing the powder injection velocity or O/F ratio in the precombustion chamber is beneficial to promoting the full combustion of magnesium powder with carbon dioxide in the combustor.When the O/F ratio in the pre-combustion chamber decreases from 0.67 to 0.31, the combustion efficiency of the combustor increases by 8% under Configuration B.The reason may be that powder injection velocity directly affect the O/F ratio and two-phase flow in the pre-combustion chamber and mixing section.Hence,the magnesium powder flow velocity plays a dominant role in the combustion characteristics of the engine.

Last, the combustion with a pre-combustion chamber is more conducive to combustion stability.At the same time,during the ignition phase, the starting of the ramjet will be more stable.The pre-combustion chamber should be at a higher combustion temperature,enabling the magnesium powder to achieve stable combustion in the combustion chamber and maintain a high combustion efficiency.Furthermore,radial injection of powders is not conducive to the uniform and stable blending of powders into mainstream carbon dioxide.Therefore, the stability of the combustion chamber pressure is easily affected by the stability of the powder injection.What’s more serious is that seriously condensed phase deposits will be formed on the inner wall of the engine opposite the powder injection port during the working process of the engine, which is not conducive to the stable operation and repeated use of the engine.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

This study was supported by the Fund of Advance Research Projects of Manned Spaceflight, China (No.050303).Thanks to Dr.Lok Han Josaih Lo for the primary revision of this paper, and to the reviewers for their constructive comments.Finally, the author would like to thank Dr.Xiaoyan Yang for her continuous support.